We’re happy to introduce results analysis, a new way to get insight into your mabl cloud runs!
Now available in early access for browser tests, results analysis uses generative AI to analyze test output and provide a succinct explanation of why the test failed. With results analysis, you can quickly understand failures, identify patterns in failures, and reduce overall time spent on identifying and resolving bugs.
Try it out
Before you start, update to the latest version of the mabl Desktop App: Help > Check for updates. During the early access period, workspace owners can enable results analysis on the Labs page: Settings > Labs.
Generating a failure analysis
On the test output page for a failing browser test, click on the Why did this test fail? button at the top of the page to generate a failure analysis.
A test failure analysis
Mabl sends the following test output data to the model to generate a failure analysis:
- Test name and description
- Test output logs and screenshot from the failing test step
- Test output logs and screenshot from the same test step in a recent passing test, if there is one
If you’re satisfied with the generated failure analysis, click on the Accept button to save it. Once saved, you can access the failure analysis from the test output page whenever you need to.
Suggested failure reasons
As part of the failure analysis, mabl includes a suggested failure reason. Your workspace's categorized failure reasons appear on the mabl dashboard, the release coverage dashboard, and in the BigQuery integration. When applied consistently and accurately to failed test runs, failure reasons can help your team understand overall application quality and save time on review.
Categorized failures on the mabl dashboard
Submit feedback
Since results analysis is available as an early access program, we highly encourage you to provide feedback on this feature in the mabl Product Portal. Your feedback helps us build a better product for you and our community!
Generative AI at mabl
Results analysis is built on top of Google Cloud's AI tools. Neither mabl nor our service partner, Google Cloud, use customer data for training these models. If you have any concerns about the use of generative AI in mabl, please reach out to your customer success manager.
For more information about how mabl handles customer data, see the article on data encryption.